remind merge requestshttps://gitlab.pik-potsdam.de/hilaire/remind/-/merge_requests2020-12-21T16:57:57+01:00https://gitlab.pik-potsdam.de/hilaire/remind/-/merge_requests/290update scripts to future proof against R 4.0.0+ updates2020-12-21T16:57:57+01:00Michaja Pehlupdate scripts to future proof against R 4.0.0+ updates - https://stat.ethz.ch/pipermail/r-announce/2020/000653.html
> R now uses a stringsAsFactors = FALSE default, and hence by
> default no longer converts strings to factors in calls to
> data.frame() and read.table().
... - https://stat.ethz.ch/pipermail/r-announce/2020/000653.html
> R now uses a stringsAsFactors = FALSE default, and hence by
> default no longer converts strings to factors in calls to
> data.frame() and read.table().
- this breaks all instances of levels() being used on data.frames
read using read.csv()
- switching to unique() solves this
- unique() used on factors (R < 4.0.0, currently on the cluster) still
returns the same values as levels(), but as a factor, which does not
cause problems (but is a little ugly in interactive use)https://gitlab.pik-potsdam.de/hilaire/remind/-/merge_requests/247add symlinks to MAGICC files2020-10-06T13:30:38+02:00Michaja Pehladd symlinks to MAGICC filesI'm still of the opinion that we can just add the MAGICC files, since you also get them on their website, but here goes:
In order to have a REMIND with on less source of constant failure (#180), we can add symlinks to the `./core/magi...I'm still of the opinion that we can just add the MAGICC files, since you also get them on their website, but here goes:
In order to have a REMIND with on less source of constant failure (#180), we can add symlinks to the `./core/magicc/` directory. The symlinks only contain the information where the files are to be found ([e.g.](https://github.com/0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q/remind/blob/feature/symlink_magicc/core/magicc/EDGAR_CH4_EMIS_LANDUSE.IN)), but no actual data. When the run is assembled by `./scripts/start/prepare_and_run.R`, the `./core/magicc/` directory [is copied](https://github.com/remindmodel/remind/blob/3e4fc335245fc37c4ed0572eb4aee960a8c29182/scripts/start/prepare_and_run.R#L403) using `file.copy()` [which copies the targets of the links, not the links](https://stat.ethz.ch/R-manual/R-devel/library/base/html/files.html):
> On a POSIX filesystem the targets of symbolic links will be copied rather than the links themselves
so we end up with proper MAGICC in the output directory, without any actuall MAGICC in the repository.
If this is agreeable, pick a proper place for the MAGICC files, not my home directory ;)https://gitlab.pik-potsdam.de/hilaire/remind/-/merge_requests/205:bug: Fix bug in function configure_cfg2020-07-20T13:18:19+02:00Jerome Hilaire:bug: Fix bug in function configure_cfg*Created by: johanneskoch94*
Hi David, is this ok? So to actually remove only files that end with ".gdx"? *Created by: johanneskoch94*
Hi David, is this ok? So to actually remove only files that end with ".gdx"?